Multi-scale wavelet algorithm based noise reduction in ultrasonic images at elevated temperatures

Abstract

Time-of-Flight Diffraction (TOFD) is an advanced ultrasonic inspection technique best suited for inspection especially inservice wherein the focus is on detection of cracks and linear indications which can propagate during service. The technique is widely used in process industries and also in strategic sector like nuclear. In the case of ISI in nuclear industry, the main difficulties likely to arise during such inspection are the high temperature (during shut down) and the background radiation. With the advent of high temperature probes, this limitation can be overcome. However, a fundamental factor which still is a cause for concern is the degradation of the signal to noise ratio due to thermal noise. Experiments conducted by the authors reveal that the contrast of acquired TOFD Images reduces progressively with increasing temperatures. As the temperature increases beyond 373K, the noise makes interpretation of images difficult and thus possibility of linear defects getting missed out arise. It is well known that the wavelets can be effectively employed for denoising. Wavelets have also been applied in general for denoising UT signals. However, its application for UT signals from stainless steel and which is at high temperature has not been attempted so far. In this paper, we explore the innovative application of wavelet thresholding for denoising applications due to its greater flexibility. Analysis of the restored image indicate that the Symlets family with the suitable thresholding has Higher Peak Signal to Noise Ratio(PSNR) value than the Coiflets wavelet family.